Xingyi Xie Assignment 3

Assignment 3.Rmd

@@ -1,4 +1,11 @@
-# Assignment 3 - Social Network Analysis
+
+---
+title: "Assignment 3 - Social Network Analysis"
+author: "Xingyi Xie"
+date: "2020/10/10"
+output: html_document
+---
+
 
 ## Part I
 Start by installing the "igraph" package. Once you have installed igraph, load the package.
@@ -22,7 +29,7 @@ So let's convert our data into an edge list!
 
 First we will isolate the variables that are of interest: comment.from and comment.to
 
-```{r}
+```{r message=FALSE, warning=FALSE}
 library(dplyr)
 
 D2 <- select(D1, comment.to, comment.from) #select() chooses the columns
@@ -75,7 +82,7 @@ VERTEX <- unique(VERTEX)
 
 Now we have both a Vertex and Edge list it is time to plot our graph!
 
-```{r}
+```{r message=FALSE, warning=FALSE}
 #Load the igraph package
 
 library(igraph)
@@ -96,7 +103,7 @@ plot(g,layout=layout.fruchterman.reingold, vertex.color=VERTEX$gender)
 
 plot(g,layout=layout.fruchterman.reingold, vertex.color=VERTEX$gender, edge.width=EDGE$count)
 
-````
+```
 
 ## Part II
 
@@ -106,6 +113,48 @@ In Part II your task is to [look up](http://igraph.org/r/) in the igraph documen
 * The vertices are colored according to major
 * The vertices are sized according to the number of comments they have recieved
 
+
+```{r message=FALSE, warning=FALSE}
+#Load the igraph package
+library(igraph)
+g <- graph.data.frame(EDGE, directed=TRUE, vertices=VERTEX)
+class(g)
+vertex.attributes(g)
+edge.attributes(g)
+summary(g)
+g1 <- igraph::graph_from_data_frame(d= EDGE,vertices = VERTEX, directed = T)
+plot(g1)
+
+# isolated nodes
+detach(package:igraph)
+library(statnet)
+library(intergraph)
+g2 <- asNetwork(g1)
+length(isolates(g2))
+plot(g2)
+plot(g,edge.width=0.1)
+plot(g,vertex.color=VERTEX$major,edge.size=0.01)
+plot(g,edge.size=0.1,vertex.color=VERTEX$major,vertex.size=EDGE$count)
+g %>%
+  plot()
+g1 %>%
+  plot()
+plot(g2)
+degree(g2)
+closeness(g2)
+# Betweenness centrality
+# B(ni) = ΣG(jk)(ni)/G(jk)
+# where G(jk) is the geodesic between nodes j and k.  
+# G(jk)(ni) is the number of geodesics between nodes j and k that contain node i.
+betweenness(g2)
+
+####Model
+Network_model <- ergm(g2 ~ edges + 
+                                nodefactor("gender")+
+                                nodefactor("major"))
+summary(Network_model)
+```
+
 ## Part III
 
 Now practice with data from our class. This data is real class data directly exported from Qualtrics and you will need to wrangle it into shape before you can work with it. Import it into R as a data frame and look at it carefully to identify problems.
@@ -117,6 +166,109 @@ Once you have done this, also [look up](http://igraph.org/r/) how to generate th
 * Betweeness centrality and dregree centrality. **Who is the most central person in the network according to these two metrics? Write a sentence or two that describes your interpretation of these metrics**
 
 * Color the nodes according to interest. Are there any clusters of interest that correspond to clusters in the network? Write a sentence or two describing your interpetation.
+```{r message=FALSE, warning=FALSE}
+dd1 <- read.csv("hudk4050-classes.csv", stringsAsFactors = FALSE, header= TRUE)
+D1 <- dd1
+dd1 <- dd1[-1,-2,]
+colnames(D1) <- D1[1,]
+```
+#Data cleaning
+#transform into long data
+
+```{r message=FALSE, warning=FALSE}
+library(tidyr)
+library(dplyr)
+library(stringr)
+library(janitor)
+D1 <- slice(D1,3:49)
+#Remove last column
+D1 <- select(D1,1:8)
+#Merge name columns
+D1 <- unite(D1,"name",`First Name`, `Last Name`, sep="")
+#Remove unpredictable characters from names
+D1$name <- str_replace(D1$name, "`", "")
+#Make all name captalized first letters only
+D1$name <- str_to_title(D1$name)
+#Make all class letters capitals
+D1 <- D1 %>% mutate_at(2:7, list(toupper))
+#Remove whitespace between letters and numbers in class
+D1 <- D1 %>% mutate_at(2:7, str_replace_all, " ", "")
+```
+
+# Data restructuring
+```{r}
+# Create a dataframe with two variables, student and class
+D2 <- D1 %>%gather(labe, class, 2:7, na.rm = TRUE, convert = FALSE) %>% select(name, class)
+#Create a new variable containing 1s that will become our counts
+D3 <- D2 %>% tabyl(name,class)
+rownames(D3) <- D3$name
+D3 <- select(D3, -name, -HUDK4050)
+```
+
+#Matrix operations
+```{r}
+#Convert to matrix
+D3 <- as.matrix(D3)
+#Create person-person matrix
+D3 <- D3 %*% t(D3)
+```
+
+#Graphing
+```{r}
+library(igraph)
+g <- graph.adjacency(D3, mode="undirected", diag = FALSE)
+plot(g,layout=layout.fruchterman.reingold,
+     vertex.size = 4,
+     vertex.label.cex =0.8,
+     vertex.label.color="black",
+     vertex.color="yellow")
+
+Network <- g %>% 
+                 simplify(remove.multiple = TRUE,remove.loops = TRUE) %>% 
+                 delete.vertices(.,which(degree(.)==0)) %>% 
+                 intergraph::asNetwork() 
+
+plot(Network)
+```
+
+
+```{r}
+# degree centrality
+degree(g)
+# Closeness centrality
+# The inverse of the sum of all the distances between node i and all the other nodes in the network.
+closeness(g)
+
+# Betweenness centrality
+# B(ni) = ΣG(jk)(ni)/G(jk)
+# where G(jk) is the geodesic between nodes j and k.  
+# G(jk)(ni) is the number of geodesics between nodes j and k that contain node i.
+betweenness(g)  
+df.prom <- data.frame(
+           deg = degree(g),
+           cls = closeness(g),
+           btw = betweenness(g))
+plot(df.prom$deg,df.prom$btw)
+which(df.prom$deg>500)
+summary(df.prom)
+which(df.prom$btw>40)
+df.prom[45,]
+df.prom <- df.prom[-45,]
+summary(df.prom)
+which(df.prom$btw>10)
+df.prom[17,]
+
+```
+
+### Answer: I think Jia Shengyu is the most central person in the network. She has the second highest betweenness, which is very important.
+
+```{r}
+#creating a variable for the number of classes so I can use it as the vertex size
+plot(g,vertex.color=factor(dd1$Q18), vertex.label=NA,main="PERSON-NETWORK")
+```
+
+#I think common interest has to do with a person's major and it is likely that students in the same major have the same classes.
+
 
 ### To Submit Your Assignment